Title: Study on personalised recommendation method of online education resources under the background of teaching reform

Authors: Hanyu Zheng

Addresses: College of Cultural and Creative Industries, Changchun University of Architecture and Civil Engineering, Changchun 130000, China

Abstract: In order to achieve the research goal of solving the problems of accuracy, recall and low F1 value of traditional online education resources personalised recommendation methods, a new personalised recommendation method of online education resources under the background of teaching reform was designed. Using the BERT model to extract learner preference vectors and feature vectors of online educational resources, an improved discrete differential evolution algorithm is designed, which is used to recall and sort online educational resource sequences. Combined with the collaborative filtering algorithm to generate recommendation sequences, personalised recommendation results of online education resources are obtained. Simulation experiments show that the accuracy rate curve of this paper method relatively flat, accuracy rate is always above 93%, the maximum recall rate is 98%, and the F1 mean is 9.67, the recommended results are reliable.

Keywords: teaching reform; online educational resources; personalised recommendation; BERT model; discrete differential evolution algorithm; collaborative filtering algorithm.

DOI: 10.1504/IJCEELL.2024.139940

International Journal of Continuing Engineering Education and Life-Long Learning, 2024 Vol.34 No.4, pp.416 - 429

Received: 20 Jun 2022
Accepted: 30 Aug 2022
Published online: 12 Jul 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article